Spatial patterns and temporal trends in stillbirth, neonatal, and infant mortality: an exploration of country-level data from 2000 to 2021.

Abstract

Background: Despite significant progress in improving child survival and health, substantial disparities persist and are being increasingly threatened by a complex and dynamic global environment. In this ecological study, we investigated spatial patterns, temporal trends, and key determinants of disparities in stillbirth, neonatal, and infant mortality rates across 195 countries from 2000 to 2021.

Methods: We sourced our data from two publicly available databases: the United Nations Children's Fund Data Warehouse for mortality indicators and the World Bank for World Development Indicators. We conducted spatial analysis to assess spatial autocorrelation and identify geographical clusters of countries and applied joinpoint regression to evaluate temporal trends in mortality indicators, reported as annual percentage change. We also used forward regression analysis to determine the primary indicators influencing stillbirth, neonatal, and infant mortality rates.

Results: The average stillbirth rate in 2021 was 10.9 per 1000 total births, a significant decrease from 16.3 per 1000 in 2000. Neonatal mortality also declined from 23.6 to 13.5 per 1000 live births during the same period, while infant mortality dropped from 45.0 to 22.5 per 1000 live births. Despite these improvements, spatial analysis showed notable positive spatial autocorrelations for stillbirth, neonatal, and infant mortality rates, indicating that high mortality rates were geographically clustered, particularly in African countries forming hot-spot clusters. Conversely, developed countries in Europe and Asia formed cold-spot clusters characterised by low mortality indicators. Some countries, identified as low-high or high-low clusters, stood out due to dissimilar mortality rates compared to their neighbours, warranting further investigation. Key determinants of mortality rates included the young-age dependency ratio, prevalence of undernourishment, the percentage of women aged 15 and older living with HIV, the incidence of tuberculosis, and the adolescent fertility rate - all of which showed a positive association with higher mortality rates. In contrast, factors such as the use of at least basic sanitation services, mean years of schooling, and government effectiveness had an inverse relationship, contributing to lower mortality rates.

Conclusions: By identifying hotspots and outliers, this study highlights the need for targeted health interventions and efficient resource allocation. This approach ensures that efforts are strategic and impactful, focussing on areas with the greatest need.